analyzing visual metaphor and metonymy to understand...
TRANSCRIPT
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 1
ORIGINAL RESEARCHpublished: 07 January 2019
doi: 10.3389/fpsyg.2018.02527
Edited by:Gerrit C. Van Der Veer,
University of Twente, Netherlands
Reviewed by:Danzhu Li,
University of Twente, NetherlandsBarbara
Lewandowska-Tomaszczyk,State University of Applied Sciences
in Konin, Poland
*Correspondence:Ryoko Uno
Specialty section:This article was submitted to
Human-Media Interaction,a section of the journalFrontiers in Psychology
Received: 14 August 2018Accepted: 27 November 2018
Published: 07 January 2019
Citation:Uno R, Matsuda E and
Indurkhya B (2019) Analyzing VisualMetaphor and Metonymy
to Understand Creativity in Fashion.Front. Psychol. 9:2527.
doi: 10.3389/fpsyg.2018.02527
Analyzing Visual Metaphor andMetonymy to Understand Creativityin FashionRyoko Uno1* , Eiko Matsuda2 and Bipin Indurkhya3
1 Division of Language and Culture Studies, Institute of Engineering, Tokyo University of Agriculture and Technology, Tokyo,Japan, 2 Division of Advanced Information Technology and Computer Science, Institute of Engineering, Tokyo Universityof Agriculture and Technology, Tokyo, Japan, 3 Department of Cognitive Science, Institute of Philosophy, JagiellonianUniversity, Kraków, Poland
The role of figurative language such as metaphor and metonymy in creativity has beenstudied in cognitive linguistics. These methods can also be applied to analyze non-linguistic data such as pictures and gestures. In this paper, we analyze fashion designby focusing on visual metaphor and metonymy. The nature of creativity in fashion designhas not been fully studied from a cognitive perspective compared to other related fieldssuch as art. We especially focus on fashion design as a communication tool betweenthe designer and audience in conveying a designer’s image of human beings. Photosfrom two fashion shows were analyzed. We carried out an experiment to compare howhuman images in two shows are interpreted by those who are familiar with fashion andthose who are not. We obtained three results: (1) As far as figurative (metaphorical andmetonymic) interpretations of human images are concerned, two groups with differentlevels of familiarity with fashion had significantly different patterns of responses to twoshows. (2) For the non-figurative interpretations (such as physical or personal attributes),no significant difference in the pattern of response to the show was observed betweenthe two groups. However, the participants as a whole responded to the two showsdifferently. (3) In addition, for the non-figurative interpretations, the fashion experts foundsignificantly more attributes in human images than the other group. The results showthat the analysis of figurative interpretations is effective in understanding how familiaritywith fashion affects the mode of seeing fashion shows.
Keywords: fashion design, creativity, cognitive linguistics, metaphor, metonymy
INTRODUCTION
An influential designer in the fashion industry today, Karl Lagerfeld, once said “Fashion is alanguage that creates itself in clothes to interpret reality.” Fashion design is an act of creating whatpeople put on, and people are always presupposed in this process. Thus, to design fashion is todesign the human image that sustains this fashion. Lagerfeld’s declaration that fashion interpretsreality also suggests that fashion proposes an image of the human whose fashion reflects reality.When we focus on fashion as a creation of human images, we make a case to examine fashionwithin cognitive science.
Frontiers in Psychology | www.frontiersin.org 1 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 2
Uno et al. Metaphor and Metonymy in Fashion
Following Lakoff and Johnson (1980), the role of figurativelanguage, especially metaphor, in creative thought has beenwidely studied. Not only verbal but also visual metaphor andmetonymy have been analyzed to reveal dynamism in conceptualstructures (Indurkhya and Ojha, 2013). This paper seeks toreveal creativity in fashion by focusing on visual metaphor andmetonymy. Compared to related fields such as art or productdesign, creativity in fashion has not received much scholarlyattention. For example, in the Cambridge Handbook of Creativity(Kaufman and Sternberg, 2010), while there are independentchapters on art (Locher, 2010) and product design (Cropley andCropley, 2010), fashion does not even appear in the index.
Not many studies of fashion creation take on a cognitiveperspective, although some have employed a sociologicalperspective. Kawamura (2005) proposed a field of study called“fashion-ology,” pointing out that fashion and clothing are twodifferent things: the former, which is invisible, is realized throughthe latter, which is visible. The origin of the term “fashion,” as itis currently used, can be traced to 1489 (Brenninkmeyer, 1963).Though fashion was limited to the upper class gentry in the15th century, it became more democratic in the 20th century.This is only one of the many aspects telling us that the meaningof fashion has shifted over years. Nonetheless, “change” and“novelty” are still regarded as two main characteristics of fashion(Kawamura, 2005).
Even in the sociological context, there is no consensus onwhether fashion can be classified as art (Miller, 2007). Moreover,Skov et al. (2009) argue that even when we see fashion as art,there are two different interpretations. On one hand, fashion canbe seen as an everyday art form. For example, Wilson (1985)describes fashion as a way to express modernity, even by ordinarypeople. On the other hand, we can see fashion designers as artists.In this perspective, the most important event in fashion is thefashion show: it is an event where the fashion designer is incomplete control of esthetic concepts. Fashion shows began byconveying the image of fashion to end customers. However, theirrole has changed since the 1960s, with the rise of fashion photosand magazines (Moeran, 2006), and the press became the mainaudience of these shows, freeing them from having to promotesales and making them autonomous (Skov et al., 2009).
In this paper, we take a cognitive scientist’s point of view tocapture the essence of fashion, i.e., the creation of human images,and follow the abovementioned position that regards the designeras an artist. We use photos from fashion shows, for fashion showsreflect the designer’s images directly, as argued by Skov et al.(2009). We understand the show as a setting to communicatehuman images from the designer to the audience through thedesign medium. In this case, the designer and the trainedaudience share creativity in fashion. In the research presentedhere, by focusing on visual metaphor and metonymy, we examinehow photos from fashion shows are variously interpreted by thosewho are familiar with fashion compared to those who are not.
Visual Metaphor, Visual MetonymyIn cognitive linguistics, the role of figurative language, especiallymetaphor, in creativity has been studied (Lakoff and Johnson,1980; Indurkhya, 1992). Metaphor is a non-literal expression that
uses a form to express a meaning that is similar to the literalmeaning, an example being “time is money.” Literally, time isnot money. The reason why time is said to be money in thisexpression is because there is a common property between them:being “valuable.” Thus, to understand the concept of time, weuse the concept of money. It is a mechanism to expand ourunderstanding of things by utilizing already acquired knowledge.Lakoff and Johnson (1980) point out that mapping the knowledgestructure of money to that of time is happening at the conceptuallevel, not the verbal level. To confirm this point, Thibodeauand Boroditsky (2011, 2013) carried out experiments to showthat by presenting different metaphors, we can affect people’schoice of actions. As Forceville (2009, 2016) notes, if metaphoris conceptual, it cannot be limited to language. Among casesof non-verbal metaphor, the most studied are those in picturesand gestures. Visual metaphor has been studied in comics,advertisements, and films but not in fashion. For example,in Japanese comics, background drawings can metaphoricallyrepresent the mental state of the protagonist (Shinohara andMatsunaka, 2009). Consider an example of visual metaphor inadvertisement from Indurkhya and Ojha (2017). In the imageshown in Indurkhya and Ojha (2017), 29, the man at the centeris wearing headphones and is listening to music at an airport.He is surrounded by people whose heads are replaced with loudspeakers. Here, the visual metaphor is formed by mapping loudspeakers to human heads. This suggests that the sound emittedby people is as loud as the sound from a loud speaker. Inother words, people are noisy. The property of a loud speaker(“emitting loud noise”) can be naturally projected onto a humanbeing (“noisy person”).
In figurative language, metaphor and metonymy are themost widely studied phenomena. Metonymy is also a non-literalexpression, but here, the shift from the literal meaning to afigurative meaning is based on contiguity, not similarity. Forexample, when we say, “We need more hands,” we mean that weneed more people who can help us, not only their hands. Wecan refer to humans as a whole by their parts such as hands orheads because they are close. Another technique of metonymyis referring to a person by what he or she is wearing. Thefamous “little red riding hood” is a good example. It refers tothe girl, not her hood. What we put on is close to us. Whilemetaphor functions by expanding our understanding, metonymyfunctions via landmarks (Langacker, 2008; Nishimura, 2008). Themetonymy can also be observed in other modalities: Catalano andWaugh (2013) analyzed both verbal and visual metonymies andshowed how they contributed to shaping public opinion.
In brief, both metaphor and metonymy are seen in visualmodality and contribute to changing the cognitive state ofpeople. However, they have not been applied to analyze humanimages in fashion shows. In the present research, our goal is toinvestigate how an expert fashion show viewer understands thefashion show differently from a non-expert viewer by focusing onfigurative (i.e., metaphorical and metonymic) and non-figurativeinterpretations. Our results show that it is only by payingattention to the figurative interpretations that we can understandthe characteristics of the way people who are familiar with fashionsee human images in fashion shows.
Frontiers in Psychology | www.frontiersin.org 2 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 3
Uno et al. Metaphor and Metonymy in Fashion
MATERIALS AND METHODS
Ethics StatementThis study was approved by the Institutional Review Board(IRB) of Tokyo University of Agriculture and Technology,Tokyo, Japan. All participants gave written informed consent inaccordance with the Declaration of Helsinki.
ParticipantsNinety-six participants took part in the experiment. Thisincluded 24 students from a private fashion design school (62.5%female) and 72 university students majoring in scientific fields(47.22% female). Henceforth, we call the former set of studentsthe fashion group and the latter the non-fashion group. Thefashion group participants were between 16 and 30 years old(average 25.65 years), and the non-fashion group participantswere between 19 and 22 years old (average 19.68 years). Thedifference in the age groups between the two sets of participantsreflects the differences between their schools. In Japan, moststudents enter undergraduate school just after finishing highschool, which is at the age of 18. Conversely, in fashionschools, students come from various backgrounds, and many
of them have been in employment. The participants wereall Japanese except for two Korean students in the fashiongroup and one Chinese student in the non-fashion group.We did not check the personal cultural backgrounds of theparticipants in the experiment, but the Japanese students in bothschools were mostly brought up in Japan. All participants wereproficient in Japanese, which ensured that they understood theinstructions. They participated in the experiment in their ownclassrooms.
StimuliTwo sets of pictures, each from different fashion shows, werechosen and shown on the question sheet (permission wasobtained from the copyright holders). In each picture, a modelwalking the runway was shown from his or her front orback. One show was the 2009–2010 Autumn Winter Collectionby writtenafterwards with the designer Yoshikazu Yamagata.Figure 1A shows the photos chosen from this show. Theother show was the 2014–2015 Autumn Winter Collection byMikio Sakabe with designers Mikio Sakabe and Shueh Jen-Fang.Figure 1B shows the photos used in the questionnaire from thisshow. Both shows were held during the Japan Fashion Week in
FIGURE 1 | Photos presented in the experiment: Photos from the two fashion shows: the garbage show (A) the letter show (B). The larger photos are the targetphotos. These images are being reproduced with the permission of the copyright holder [(A) writtenafterwards inc; (B) Fashionsnap.com]. The original sourcesappeared in (A) http://www.writtenafterwards.com; (B) https://www.fashionsnap.com/.
Frontiers in Psychology | www.frontiersin.org 3 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 4
Uno et al. Metaphor and Metonymy in Fashion
Tokyo. All three designers teach at the private fashion schoolwhere the fashion group participants studied.
Though the designers featured in the two sets of picturesmainly sell their products in Japan, they are internationallyrecognized. For example, Yamagata and Sakabe were shortlistedfor the LVMH Prize in 2015 and 2016, respectively. The LVMHprize is launched by LVMH Moët Hennessy Louis Vuitton SEin order to honor young fashion designers around the world. Inaddition, all three designers were educated in Europe. Yamagatastudied at Central Saint Martins College of Arts and Design(current Central Saint Martins, University of the Arts London) inthe United Kingdom, Sakabe at the Royal Academy of Fine Arts,Antwerp, Belgium, and Shueh at La Cambre in Brussels, Belgium.Yamagata and Sakabe both graduated at the top of their respectiveclasses.
Each set contained a target photo: the participants were askedto describe the human image in the photo. A further three photosin the set were presented to the participants so as to contextualizethe target photo. We used three criteria to choose the targetpictures. First, the show must target displays of the human imageand not just clothes. Second, since the current research analyzesthe distribution of metaphorical and metonymic interpretations,the picture must be open to both kinds of interpretations. Third,two target photos must be distinctly different so as to determinewhether the participants responded differently to them.
We chose two pictures that met these criteria as target pictures.To begin with, Yamagata and Sakabe (2013), who were thedesigners of the target photos, explicitly mentioned in a bookthey co-authored that they present human images at fashionshows: on the cover of this book is written, “Beyond makingclothes, we make a new human.” Moreover, these picturescan be interpreted both metaphorically and metonymically.Metonymic interpretations of clothes are not difficult because thecloseness between the clothes and the person wearing them isapparent. However, not many fashion photos can easily inducemetaphorical interpretation. In the first target photo (the largephoto in Figure 1A), instead of ordinary clothes, the modelis wearing some garbage that was originally collected fromarchitecture departments at universities. We can say that thegarbage replaces the clothes, just like speakers replace heads asin the example presented before. In the second target photo (thelarge photo in Figure 1B), we can say that instead of clothes, themodel is wearing a letter on his back. This is a Chinese letterused in Japanese and means “love.” Finally, the two shows aredistinctive. When we compare garbage and a letter, the former isa substance (or a texture), and the latter is an object. In addition,the former is concrete, and the latter is abstract.
Henceforth, we call the former show the garbage show andthe latter show the letter show. Among the participants, only onefrom the non-fashion group knew of the garbage show before theexperiment. No one knew the letter show. Conversely, among thefashion group, twelve participants knew about both shows, fourknew only the garbage show, and one knew only the letter show.
ProcedureThe participants were asked to answer two questions on thequestionnaire by writing freely. Each side of the paper had four
photos from one of the two collections, as shown in Figure 1.The photo images were all in color. Two types of sheets wereprepared: the first type presented the garbage show first, while theother presented the letter show first. We randomly assigned eitherone of the sheets to the participants. For each set of pictures,we asked: “The following four pictures are from a fashion show.Look at the third large photo and answer what kind of humanis represented by this photo.” We had additional informationon the questionnaire for the participants. For the garbage showphotos, there was an explanation that the material used for theclothes was garbage. For the letter show photos, there was anexplanation that read that what was on the back of the model wasa letter.
RESULTS
Ratings for Figurative and Non-figurativeInterpretationsAfter all the data were collected, the answers (freely writtendescriptions of human images represented by the two targetphotos) were rated in six categories by three coders who workedseparately. There were two groups and six types of categories torate. The first group included categories that showed how thefigurative interpretations of the human images were working.The figurative group included the categories “metonymic” and“metaphorical.” These were the two main categories observed inthis research. To determine how figurative interpretations work,we also put together the non-figurative interpretations group.This second group comprised four categories characterizinghuman personality: physical, personal, relational, and social role.This categorization was based on Smilek et al. (2007), whoanalyzed the characteristics of a personality description of anobject by a participant with synesthesia. The details of the ratingsby each coder are shown in the Supplementary Material.
Among the answers by the participants, those that did notdescribe a human image at all were excluded: for example,answers such as “I don’t know” or comments about the designer.For the remaining answers, the coders were asked to rate howwell each answer (the description of human image) matched theattributes in the six categories. The rating was carried out using asix-point Likert scale (1, matches completely; 6, does not match atall).
Each category was explained to the coders via an instructionsheet. The descriptions of the six categories provided to thecoders are given below. Some examples from the participants’answers are also shown below after the descriptions of eachcategory. These eight examples were all recognized by the codersto generally match their respective categories. Two coders gave a1, and one coder gave a 2 for the answers in (2, 3, and 4). Thecoders rated all the remaining examples as 1.
For the transcription of the Japanese examples, the Hepburnsystem of Romanization was used. The following abbreviationswere used in the word-to-word translations: ACC for accusative,COMP for complementizer, GEN for genitive, NOM fornominative, PASS for passive, PAST for past tense, and POL forpolite.
Frontiers in Psychology | www.frontiersin.org 4 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 5
Uno et al. Metaphor and Metonymy in Fashion
For the metonymic category, the instruction sheet for thecoders instructed that the answer matches this category when itexpresses a person who is close (or contiguous) to garbage ora letter. It is also noted that this “closeness” is physical, suchas spatiotemporal proximity or causation. Thus, closeness doesnot include similarity. Another noteworthy point is that for theletter collection, not only should we consider the closeness to theletter but also the closeness to the concept of “love” (which is themeaning of the letter). This category includes someone who hasgarbage or the letter (or the concept of love) on his or her bodyor someone who is staying close to garbage or the letter (or theconcept of love), etc. Example (1) expresses a person who is closeto garbage and (2) expresses a person who is close to the letter.
(1) Metonymictsukaeru mono wa hirot-teshimau youna hito to kanji-masuuseful thing TOP pick.up-PERF like person COMP feel-POL“I feel that she is a kind of person who picks up whatever shefinds useful.”
(2) Metonymicmukashi no bushoh no kabuto no ai noold.days GEN warrior GEN helmet GEN love GENimeeji ga tsuyoku isamashii hito ni kanji-taimage NOM strong gallant person as feel-PAST“The letter ‘ai ( = love)’ strongly reminds me of the letteron the warrior’s helmet in old days, so I felt him as a gallantperson.”
The second example requires some explanation. In old Japan,military commanders wore a helmet with a Chinese character onthe front as a part of the design. One of the famous commanderswore a helmet with the letter ai. The answer in (2) refers to thishelmet.
Next, for the metaphorical category, the coders wereinstructed that the answer would match the attribute in themetaphorical category when it expressed a person who is similarto garbage or the letter (or the concept of love). The answer in (3)is an example that was judged by a coder to describe a person asbeing similar to garbage, i.e., someone who can be abandoned bysociety.
(3) Metaphoricalshakai no haguruma to-shite tsukai sute-rareru hitosociety GEN cog as use discard-PASS person“A person who is used and discarded as a cog in a wheel insociety.”
The following example (4) was judged by a coder to describe aperson who is similar to the letter:
(4) Metaphoricalshuchoh no tsuyoi hitoassersion GEN strong person“An assertive person.”
For the non-figurative group, we followed the categorizationused in the study by Smilek et al. (2007), which analyzedthe personification of inanimate objects by a participant with
synesthesia. The instruction sheet provided some examples ofeach category to the coders as follows: Physical attributes includefeatures such as male, in 40s, black hair, Japanese, etc. Personalattributes include serious, optimistic, mischievous, etc. Relationalattributes include friendly, dominant, etc. Finally, social roleattributes include younger brother, god, leader, etc. The followingare some examples judged by the coders as belonging to eachcategory:
(5) Physicalkireina onna-no-hitobeautiful woman“A beautiful woman.”
(6) Personalkurai hitodark person“A gloomy person.”
(7) Relationalmawari to no kyohchoh yori-mopeople.around with GEN cooperation rather.thanjishin no ishi o sonchoh-suru kanji ga shi-masu.oneself GEN will ACC respect feeling NOM do-POL“I feel that this person respects her will more thancooperating with people around her.”
(8) Social roleyakuzagangster“A gangster.”
After all the answers were rated by the coders, two procedureswere followed before the analysis. First, we converted six-pointratings into binary values. Ratings 1–3, which were affirmative tothe matching, were replaced with the value 1. Ratings 4–6, whichwere negative evaluations of the matching, were replaced with thevalue 0. Next, we merged the three ratings (one by each coder)for each answer as follows. When there was a 1 by any coder in agiven category, the final rating was set to 1. Only when the ratingsby all the three coders were 0 was the final rating set to 0.
Analysis of Figurative InterpretationsIn this part, we focus on the figurative interpretations, namely,metonymic and metaphorical categories. As noted above, theanswers to each question were converted into binary values foreach of the two categories (metonymic or metaphorical). Werefer to these binary values as “category degrees” for furtheranalysis.
Taking the category degree for the objective variable, a three-way ANOVA was performed for the photos (garbage or letter),participant groups (fashion or non-fashion), and category types(metonymic or metaphorical). There were significant interactionsamong the photos, the participant groups, and the categorytypes [α = 5%, F(1, 370) = 9.04, p = 0.0028], and between thephotos and the category types [F(1, 370) = 4.11, p = 0.043]. Nosignificant interactions were observed between the photos and
Frontiers in Psychology | www.frontiersin.org 5 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 6
Uno et al. Metaphor and Metonymy in Fashion
the participant groups [F(1, 370) = 1.35, p = 0.25] or betweenthe participant groups and the category types [F(1, 370) = 2.40,p = 0.12]. Therefore, we analyzed the data from the fashionand non-fashion participant groups separately so as to find thesub-effects between the photos and category types. For eachparticipant group, a two-way ANOVA was conducted for thephotos and category types.
Regarding the fashion group, there was a significantinteraction between the photos and the category types [α = 5%,F(1,88) = 9.2933, p = 0.0030]. We therefore proceeded to analyzethe simple main effects of the interaction. Figure 2 showsthe mean category degree for metonymic and metaphoricalcategories, respectively (Figure 2 upper: metonymic, lower:metaphorical). The x-axis denotes photos (garbage or letter),and the y-axis indicates the mean category degree. The errorbars denote SEM. The results of the simple main effectsshow a significant difference for the metonymic category [F(1,88) = 9.6561, p = 0.0025] but not for the metaphorical category[F(1, 88) = 1.4491, p = 0.2319], which are summarized in Figure 2by the notation, ∗∗∗p < 0.01, n.s. p ≥ 0.05.
Regarding the non-fashion group, the mean category degreefor each statistical group is depicted in Figure 1B. There wasneither any significant interaction between the photos andcategory types [α = 5%, F(1, 282) = 0.9567, p = 0.3289] nor maineffects in the photos [F(1, 282) = 0.0096, p = 0.9222] and categorytypes [F(1, 282) = 3.748, p = 0.0539].
Analysis of Non-figurative InterpretationsIn this section, we focus on the remaining categories reflectedin the non-figurative interpretations. The categories were
physical, personal, relational, and social role. Similar to thefigurative interpretations, we used the category degree for ouranalysis.
Taking the category degree for the objective variable, a three-way ANOVA was performed on the photos (garbage or letter),participant groups (fashion or non-fashion), and category types(physical, personal, relational, and social role). There were nosignificant interactions among the photos, the participant groups,and the category types [α = 5%, F(3, 740) = 0.72, p = 0.54],between the photos and the participant groups [F(1, 740) = 0.39,p = 0.53], or between the participant groups and the categorytypes [α = 5%, F(3, 740) = 1.80, p = 0.15]. A significant interactionwas observed between the photos and the category groups [F(3,740) = 8.03, p < 0.0001]. Significant main effects were observedin the photos [F(1, 740) = 6.4678, p = 0.011], the participantgroups [F(1, 740) = 16.86, p < 0.0001], and the category types[F(3, 740) = 81.58, p < 0.0001].
We therefore focused on the simple interaction between thephotos and the category degrees without taking the participantgroups into account. Figure 3 depicts the mean category degreefor each category (a: physical, b: personal, c: relational, d: socialrole). The x-axis denotes photos (garbage or letter), and the y-axisindicates the mean category degree. The error bars denote SEM.The results of the simple main effects show a significant differencefor the physical category [α = 5%, F(1,740) = 5.4153, p = 0.0202],the personal category [F(1,740) = 3.9691, p = 0.0467], and therelational category [F(1, 740) = 20.3284, p < 0.0001], but notfor the social role category [F(1, 740) = 0.8327, p = 0.3618]. Theresults are summarized in Figure 3 by the notation, ∗∗∗p < 0.01,∗p < 0.05, n.s. p ≥ 0.05.
FIGURE 2 | Mean tag degree across all participants. The x-axis denotes the photo (garbage or letter), while the y-axis indicates the mean category degree. The errorbars indicate SEM. The results of two-way ANOVA are summarized by the notation ∗∗∗p < 0.01, n.s. p ≥ 0.05. (A) Results from the fashion group. (B) Results fromthe non-fashion group.
Frontiers in Psychology | www.frontiersin.org 6 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 7
Uno et al. Metaphor and Metonymy in Fashion
FIGURE 3 | Mean tag degree across all participants. The x-axis denotes thephoto (garbage or letter), while the y-axis indicates the mean category degree.The error bars indicate SEM. Each sub-figure shows the results for eachcategory: (A), physical; (B), personal; (C), relational; and (D), social role. Theresults of two-way ANOVA are summarized by the notation ∗∗∗p < 0.01,∗p < 0.05, n.s. p ≥ 0.05.
We further analyzed the main effects of the participant groups.Figure 4 depicts the difference in category degree across theparticipant groups.
DISCUSSION
While creativity in art and design has been studied in cognitivescience, fashion design has received little attention. In this study,by applying studies in visual metaphor and metonymy, weinvestigated the creativity of people familiar with fashion designin watching fashion shows.
We analyzed the interpretations of the human imagesrepresented in photos from fashion shows presented to twogroups: people with and without knowledge of fashion.When we focused on figurative (metaphorical and metonymic)interpretations of these photos, we found some key differences
FIGURE 4 | Mean tag degree across all participants. The x-axis denotes theparticipant group (fashion or non-fashion), while the y-axis indicates the meantag degree. The error bars indicate SEM. The results of two-way ANOVA aresummarized by the notation ∗∗∗p < 0.01, n.s. p ≥ 0.05.
in how the two groups responded to the two shows. Thefashion group had more metonymic interpretations for theletter show, but there was no significant difference betweenthe metaphoric and metonymic interpretations for the non-fashion group. Regarding the non-figurative interpretations, nosignificant difference was found between the two groups ininterpreting the two shows. Overall, the participants found moreouter attributes in the garbage show photos and more innerattributes (personal and relational) in the letter show photos. Inaddition, the fashion group saw more non-figurative attributes inboth shows compared to the non-fashion group.
Our general conclusion is that to see how people who arefamiliar with fashion understand photos from fashion shows,simply analyzing the straightforward attributes (non-figurativeinterpretations) of the depiction of humans in fashion imagesis not enough. The distinctive feature of the fashion group inunderstanding a fashion show is reflected in the metaphorical andmetonymic interpretations. The core of figurative interpretationis how people derive new meanings from a form (Indurkhya,2007). It is a way to have rich meaning with a limited numberof forms.
More specifically, we argue that the two shows differsignificantly in the following ways. First, with respect to figurativeinterpretations, the fashion group had more metonymicinterpretations for the letter collection. We suggest that this isbecause the fashion group focused on the shape of the letterand the texture of the garbage. The difference between theshape and the texture is reflected in the language system as thedifference between countable and uncountable nouns or thedifference between nouns in English and Japanese (Nomura,1996). Moreover, some studies have shown that young childrenwho just acquired a language are sensitive to the differencesbetween shape and texture (Soja et al., 1991), and based on theacquired language, there is a difference in the importance offeatures (Imai and Gentner, 1997). It is possible that people whoare familiar with fashion are more sensitive to shape.
Frontiers in Psychology | www.frontiersin.org 7 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 8
Uno et al. Metaphor and Metonymy in Fashion
Second, in the non-figurative interpretations, the participantscharacterized the garbage show photos with outer traits andthe letter show photos with inner traits. We suggest that thisdifference emanates from the concrete versus abstract contrastbetween the two shows rather than the texture versus shapecontrast.
Needless to say, this research has some limitations. Forexample, by using four pictures from each show, we were ableto capture only some parts of the information related to fashiondesign in a fashion show. Also, since we only targeted one fashionschool, where the focus is on creative aspects of fashion, we wereunable to ascertain what would have happened had we includedparticipants from other fashion schools that teach more practicalor commercial aspects of fashion. These points should be takeninto consideration in the following studies.
For future research, we plan to study two points: (1) how theshape versus texture and the abstract versus concrete contrastswork; (2) whether the different responses to the shows by thefashion group with figurative interpretations were related tohaving more non-figurative interpretations.
Finally, we would like to make two comments based on ourresults from a more global perspective. One is from a human-agent interaction perspective and the other is a perspective fromcommunication studies.
How people interpret a rich human image from the photoof a human, i.e., without interacting with the human, seems toshare the same mechanism as how people see a non-human asa human (personification). Many researchers have explored whywe personify computers (Nass, 1996; Colburn and Shute, 2008;Nass and Yen, 2010), robots, and androids (Straub et al., 2010).Our results suggest that by manipulating clothes or appearancesas well as by training the human who interacts with agents, wecan make agents appear more humanlike.
From the perspective of human communication, the targetof this analysis is interesting because while we saw fashionshows as a form of communication between the designer andthe audience in conveying human images, the purpose of thiscommunication was not to convey the image in a precise manner.The purpose was communication itself, which was done bysharing the human image. Even with natural languages, wedo not always try to convey messages precisely. For example,when we are playing with a language (Crystal, 1998), we aresimply communicating for the sake of communicating. Thistype of activity can often lead to the creation of new words
(Schmid, 2008, 2016). However, most linguistic analyses focuson communication as an exchange of information. Even in theframework of experimental semiotics (Galantucci, 2005; Steels,2006; Kirby et al., 2008; Scott-Phillips and Kirby, 2010; Galantucciand Garrod, 2012), where the aim is to understand the emergenceof communication or language by observing the usage of artificialcommunication tools, the main focus is on communication ora language that delivers information (Uno et al., 2012). AsClark (1997) points out, humans can extend their cognitivecapacity using artifacts such as mobile phones, search engines, orlanguage. Fashion is also an artifact for us to extend ourselves.By analyzing metaphor and metonymy in fashion, we can arriveat clues to analyze fashion as a phenomenon that reflects ourcommunication capacity. We can expand our study of fashiondesign to reveal the mechanisms underlying our communicativeabilities in general.
AUTHOR CONTRIBUTIONS
RU, EM, and BI conceived the experiments. RU performedthe experiments. EM carried out the data analyses. All authorsdiscussed and interpreted the results and contributed to drafts ofthis manuscript.
FUNDING
This work was supported by MEXT/JSPS Grant-in-Aid forScientific Research on Innovative Areas #4903 (Evolinguistics)grant no. JP18H05062.
ACKNOWLEDGMENTS
We would like to thank Daisuke Nonaka, Kazuko Shinohara, andShiro Yano for helpful comments.
SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be foundonline at: https://www.frontiersin.org/articles/10.3389/fpsyg.2018.02527/full#supplementary-material
REFERENCESBrenninkmeyer, I. (1963). The Sociology of Fashion. Paris: Westdeutscher Verlag.Catalano, T., and Waugh, L. R. (2013). The language of money: how verbal and
visual metonymy shapes public opinion about financial events. Int. J. Lang.Stud. 7, 31–60.
Clark, A. (1997). Being There: Putting Brain, Body and Mind Together. Cambridge:MIT Press. doi: 10.1016/j.jal.2008.09.005
Colburn, T. R., and Shute, G. M. (2008). Metaphor in computer science. J. Appl.Logic 6, 526–533.
Cropley, D., and Cropley, A. (2010). “Functional relativity: ‘products’ and thegeneration of effective novelty,” in The Cambridge Handbook of Creativity, edsJ. C. Kaufman and R. J. Sternberg (Cambridge: Cambridge University Press),301–317.
Crystal, D. (1998). Language Play. Harmondsworth: Penguin.Forceville, C. (2009). “Non-verbal and multimodal metaphor in cognitivist
framework: agendas for research,” in Multimodal Metaphor, eds C. Forcevilleand E. Urios-Aparisi (Berlin: Mouton de Gruyter), 19–35. doi: 10.1515/9783110215366
Forceville, C. (2016). “Pictorial and multimodal metaphor,” in HandbuchSprache im Multimodalen Kontext The Language in Multimodal ContextsHandbook, eds N. M. Klug and H. Stöckl (Berlin: Mouton de Gruyter),241–260.
Galantucci, B. (2005). An experimental study of the emergence ofhuman communication systems. Cogn. Sci. 29, 737–767. doi: 10.1207/s15516709cog0000_34
Galantucci, B., and Garrod, S. (2012). Experimental Semiotics. Amsterdam: JohnBenjamins Publishing Company. doi: 10.1075/bct.45
Frontiers in Psychology | www.frontiersin.org 8 January 2019 | Volume 9 | Article 2527
fpsyg-09-02527 December 24, 2018 Time: 18:6 # 9
Uno et al. Metaphor and Metonymy in Fashion
Imai, M., and Gentner, D. (1997). A cross-linguistic study of early word meaning:universal ontology and linguistic influence. Cognition 62, 169–200. doi: 10.1016/S0010-0277(96)00784-6
Indurkhya, B. (1992). Metaphor and Cognition. Dordrecht: Kluwer AcademicPublishers. doi: 10.1007/978-94-017-2252-0
Indurkhya, B. (2007). “Creativity in interpreting poetic metaphors,” in NewDirections in Metaphor Research, ed. T. Kusumi (Tokyo: Hituzi Syobo),483–501.
Indurkhya, B., and Ojha, A. (2013). An experimental study on the role of perceptualsimilarity in visual metaphors.Metaphor Symbol 28, 233–253. doi: 10.1111/cogs.12056
Indurkhya, B., and Ojha, A. (2017). Interpreting visual metaphors: asymmetry andreversibility. Poetics Today 38, 93–121. doi: 10.1215/03335372-3716240
Kaufman, J. C., and Sternberg, R. J. (eds) (2010). The CambridgeHandbook of Creativity. Cambridge: Cambridge University Press.doi: 10.1017/CBO9780511763205
Kawamura, Y. (2005). Fashion-ology: An Introduction to Fashion Studies. Oxford:Berg.
Kirby, S., Cornish, H., and Smith, K. (2008). Cumulative cultural evolution inthe laboratory: an experimental approach to the origins of structure in humanlanguage. Proc. Natl. Acad. Sci. U.S.A. 105, 10681–10686. doi: 10.1073/pnas.0707835105
Lakoff, G., and Johnson, M. (1980). Metaphors We Live By. Chicago: University ofChicago Press.
Langacker, R. W. (2008).Cognitive Grammar: A Basic Introduction. Oxford: OxfordUniversity Press. doi: 10.1093/acprof:oso/9780195331967.001.0001
Locher, P. J. (2010). “How does a visual artist create an artwork?,” in The CambridgeHandbook of Creativity, eds J. C. Kaufman and R. J. Sternberg (Cambridge:Cambridge University Press), 131–144.
Miller, S. (2007). Fashion as art; is fashion art? Fashion Theory 11, 25–40. doi:10.2752/136270407779934551
Moeran, B. (2006). More than just a fashion magazine. Curr. Sociol. 54, 725–744.doi: 10.1177/0011392106066813
Nass, C. (1996). The Media Equation: How People Treat Computers, Television,and New Media like Real People and Places. Cambridge: Cambridge UniversityPress.
Nass, C., and Yen, C. (2010). The Man Who Lied to His Laptop: What MachinesTeach Us About Human Relationships. New York: Current.
Nishimura, Y. (2008). Metonymy underlying grammar. ENERGEIA 33, 15–26.Nomura, M. (1996). The ubiquity of the fluid metaphor in Japanese: a case study.
Poetica 46, 41–75.Schmid, H. J. (2008). New words in the mind: concept-formation and
entrenchment of neologisms. Anglia 126, 1–36. doi: 10.1515/angl.2008.002
Schmid, H. J. (2016). “Entrenchment, salience, and basic levels,” in The OxfordHandbook of Cognitive Linguistics, eds D. Geeraerts and H. Cuyckens (Oxford:Oxford University Press), 117–138.
Scott-Phillips, T. C., and Kirby, S. (2010). Language evolution in the laboratory.Trends Cogn. Sci. 14, 411–417. doi: 10.1016/j.tics.2010.06.006
Shinohara, K., and Matsunaka, Y. (2009). “Pictorial metaphors of emotion inJapanese comics,” in Multimodal Metaphor, eds C. Forceville and E. Urios-Aparisi (Berlin: Mouton de Gruyter), 265–293.
Skov, L., Skjold, E., Moeran, B., Larsen, F., and Csaba, F. F. (2009). The FashionShow as an Art form. Creative Encounters Working Paper No. 32. Copenhagen:Copenhagen Business School, 1–33.
Smilek, D., Malcolmson, K. A., Carriere, J. S., Eller, M., Kwan, D., and Reynolds, M.(2007). When “3” is a jerk and “E” is a king: personifying inanimate objects insynesthesia. J. Cogn. Neurosci. 19, 981–992. doi: 10.1162/jocn.2007.19.6.981
Soja, N. N., Carey, S., and Spelke, E. S. (1991). Ontological categories guide youngchildren’s inductions of word meaning: object terms and substance terms.Cognition 38, 179–211. doi: 10.1016/0010-0277(91)90051-5
Steels, L. (2006). Experiments on the emergence of human communication. TrendsCogn. Sci. 10, 347–349. doi: 10.1016/j.tics.2006.06.002
Straub, I., Nishio, S., and Ishiguro, H. (2010). “Incorporated identity in interactionwith a teleoperated android robot: a case study,” in Proceedings of the 19thIEEE InternationalWorkshop on Robot andHuman Interactive Communication,(Viareggio), doi: 10.1109/ROMAN.2010.5598695
Thibodeau, P. H., and Boroditsky, L. (2011). Metaphors we think with: the role ofmetaphor in reasoning. PLoS One 6:e16782. doi: 10.1371/journal.pone.0016782
Thibodeau, P. H., and Boroditsky, L. (2013). Natural language metaphors covertlyinfluence reasoning. PLoS One 8:e52961. doi: 10.1371/journal.pone.0052961
Uno, R., Suzuki, K., and Ikegami, T. (2012). An experimental approach to speakers’perspectives in a paired wall game. Adv. Complex Syst. 15:11500020. doi: 10.1142/S0219525911500202
Wilson, E. (1985). Adorned in Dreams. Fashion and Modernity. London: Virago.Yamagata, Y., and Sakabe, M. (2013). Fashion is Magic. Tokyo: Asahi Press.
Conflict of Interest Statement: The authors declare that the research wasconducted in the absence of any commercial or financial relationships that couldbe construed as a potential conflict of interest.
Copyright © 2019 Uno, Matsuda and Indurkhya. This is an open-access articledistributed under the terms of the Creative Commons Attribution License (CC BY).The use, distribution or reproduction in other forums is permitted, provided theoriginal author(s) and the copyright owner(s) are credited and that the originalpublication in this journal is cited, in accordance with accepted academic practice.No use, distribution or reproduction is permitted which does not comply with theseterms.
Frontiers in Psychology | www.frontiersin.org 9 January 2019 | Volume 9 | Article 2527